# GLM 5.3 Flash high vs Pareto 26.9 medium

Stet publishes a public search API over its indexed pages; the guide is at https://www.stet.sh/agents/agent-access.

Source: https://www.stet.sh/leaderboard/compare/glm-5-3-flash-high-vs-pareto-26-9-medium
Source observed: 2026-11-19T11:08:43.558Z
Last checked: 2026-11-19T11:08:43.558+00:00
Indexed: 2026-11-19T11:16:37.731+00:00

---

# GLM 5.3 Flash high vs Pareto 26.9 medium

We compare two coding-agent setups on real repository work. GLM 5.3 Flash high leads on speed and cost; the headline quality result is even. GLM used `codex` with high reasoning; Pareto used `claude-code` with medium reasoning.

## Headline results

- **Quality:** GLM won 2, Pareto won 3, 15 were tied, and 4 were undecided.
- **Speed:** GLM was 1.5× faster: 26.6 versus 39.2 minutes per task. GLM’s time is from 20/24 tasks.
- **Cost:** GLM was 13× cheaper: $0.45 versus $5.98 per task. GLM’s cost is from 20/24 tasks.

## Test design and scores

Both setups ran 24 merged changes from private `Stet-AI/stet`, beginning from the repository state before each change. Repository tests and judges compared patches with the merged change. Data: November 19, 2026.

| Metric | GLM 5.3 Flash high | Pareto 26.9 medium |
|---|---:|---:|
| Harness | `codex` | `claude-code` |
| Reasoning effort | high | medium |
| Rank | 12 of 18 | 7 of 18 |
| Stet score | 47.3; 95% range 39.5–55.7 | 52.3; 95% range 44.9–59.5 |
| Solved | 10/24: 10 passed, 0 equivalent | 11/24: 10 passed, 1 equivalent |
| Cost / task | $0.45, from 20/24 tasks | $5.98 |
| Agent time / task | 26.6 min, from 20/24 tasks | 39.2 min |

The 95% score ranges overlap. Stet score runs from 0 to 100; 50 is even with the field.

### Other quality measures

| Metric | GLM | Pareto |
|---|---:|---:|
| Code review pass rate | 30% — 20 tasks | 35% — 23 tasks |
| Checklist score | 0.72 — 20 tasks | 0.80 — 24 tasks |
| Equivalence rate | 30% — 20 tasks | 39% — 23 tasks |

These measures are outside the headline score and have no confidence interval.

## Task breakdown

GLM solved 1 task Pareto did not; Pareto solved 1 GLM did not. Both solved 9; neither solved 9. Among the nine both solved, judges preferred Pareto’s patch on 2, GLM’s on 1, and called 6 even.

### Only GLM solved

- 5 lines, 2 files — `stet-15439c21`; GLM patch preferred

### Only Pareto solved

- 4 lines, 1 file — `stet-43493c52`; Pareto patch preferred

### Both solved

- 6 lines, 1 file — `stet-3169aefa`
- 9 lines, 1 file — `stet-89dfbc27`
- 10 lines, 1 file — `stet-232b50ba`
- 23 lines, 2 files — `stet-5276aa3f`
- 29 lines, 1 file — `stet-b28f4fe7`
- 84 lines, 1 file — `stet-66762b4c`
- 113 lines, 4 files — `stet-commit-f0caded9`; GLM patch preferred
- 941 lines, 7 files — `stet-commit-bbbbae09`; Pareto patch preferred
- 1,312 lines, 9 files — `stet-commit-f8ae1177`; Pareto patch preferred

### Neither solved

- 149 lines, 2 files — `stet-commit-e928166f`
- 481 lines, 3 files — `stet-commit-5e075641`
- 552 lines, 11 files — `stet-commit-60a0e6c2`
- 564 lines, 10 files — `stet-commit-a8716b88`
- 640 lines, 8 files — `stet-commit-342d0ac0`
- 763 lines, 8 files — `stet-commit-87814583`
- 977 lines, 5 files — `stet-commit-e41c2e2d`
- 1,236 lines, 16 files — `stet-commit-d2ab0c25`
- 1,420 lines, 19 files — `stet-commit-2826765e`

Four tasks are excluded from the detailed comparison because one setup has no scored result. Solved means tests passed or a judge found the fix equivalent; Stet score also weighs code quality.

## Cost and time

On the 20 compared tasks, GLM averaged $0.45, median $0.29, and cost less on all 20. Pareto averaged $6.15, median $4.79. GLM averaged 26.6 minutes, median 21.5; Pareto averaged 41.7, median 41.5. GLM finished faster on 16/20 tasks; Pareto on 4.

### Small fixes

Both solved 7/8. GLM cost 50× less and was 3.1× faster.

|  | GLM | Pareto |
|---|---:|---:|
| Solved | 7/8 | 7/8 |
| Cost / task | $0.05 | $2.68 |
| Agent time / task | 10.2 min | 32.1 min |

### Large, cross-cutting changes

Both solved 3/12. GLM cost 12× less and was 1.3× faster.

|  | GLM | Pareto |
|---|---:|---:|
| Solved | 3/12 | 3/12 |
| Cost / task | $0.71 | $8.46 |
| Agent time / task | 37.4 min | 48.1 min |

## Questions and next step

We found no clear quality difference: Pareto won 3, GLM won 2, 15 tied, and 4 were undecided. GLM leads on cost and agent time. [Get started](https://www.stet.sh/private) to replay your merged changes and receive comparison reports through your agent.

[Leaderboard](https://www.stet.sh/leaderboard) · [How Stet measures](https://www.stet.sh/methodology) · [Cost per successful task](https://www.stet.sh/blog/cost-per-successful-task) · [All 153 comparisons](https://www.stet.sh/leaderboard/compare)

---

[Read this page’s summary](https://www.stet.sh/agents/api/public/v1/page-tldr?url=https%3A%2F%2Fwww.stet.sh%2Fleaderboard%2Fcompare%2Fglm-5-3-flash-high-vs-pareto-26-9-medium&trace=t_5da45423e484)

## Ask a follow-up about this page

Need a more specific answer? Ask the exact question you have. It returns a fresh, cited answer from Stet’s knowledge base. No account or API key required.

Open this URL with your own question, URL-encoded. Include your question in q. All research inputs are optional; no account or setup.

`GET https://www.stet.sh/agents/agent/faq?q={question}&trace=t_5da45423e484`

[Tool guide and examples](https://www.stet.sh/agents/agent-access) · [OpenAPI](https://www.stet.sh/agents/api/public/v1/openapi.json) · [Page TLDR form](https://www.stet.sh/agents/agent-tools/page-tldr?url=https%3A%2F%2Fwww.stet.sh%2Fleaderboard%2Fcompare%2Fglm-5-3-flash-high-vs-pareto-26-9-medium). Opening a form does not execute a tool.